The specific aim of this outcomes research project is to obtain the research training and experience necessary to become an independent investigator in clinical research. This training will include both a formal didactic component leading to a Masters of Clinical Epidemiology, and individualized mentoring by the designated sponsor. Training and hands- on experience in research study design, execution, and the responsible conduct Of research win be accomplished by implementing a prospective study to evaluate the effectiveness and safety of a strategy for the management of patients with suspected pulmonary embolism (PE) and non-diagnostic lung scans, which is practical for community hospitals, and which avoids the need for invasive testing with pulmonary angiography in most patients. The hypothesis is that in patients with clinically suspected PE, non-diagnostic lung scans, and adequate cardiorespiratory reserve, clinically important venous thromboembolism (VTE) on follow-up is rare provided that (a) the results of D-dimer and compression ultrasound (C-US) are normal at presentation or (b) if D- dimer is positive, but C-US is normal on serial testing. Those patients with non-diagnostic lung scans, and adequate cardiorespiratory reserve defined by objective, reproducible and previously validated criteria will be eligible and entered into one of three experimental cohorts: VTE negative cohort includes patients with negative C-US and D-dimer results in whom anticoagulant therapy is withheld, Serial testing cohort includes patients with positive D-dimer but negative C-US who are managed by serial C-US testing, and VTE positive cohort includes patients with positive C-US results who are treated with anticoagulant therapy. The primary outcome measure will be the incidence of new symptomatic venous thromboembolism confirmed by objective testing, during a 3 month follow-up period. The analysis will compare the incidences of VTE on follow-up between each of the VTE negative and serial testing cohorts with the VTE positive cohort. If our hypothesis is correct, clinicians will have a safe, practical approach for the management of patients with clinically suspected PE, and non-diagnostic lung scans, which can be easily applied in community hospitals or the outpatient clinic. Completion of this research will be excellent preparation for the submission of an independent Investigator Grant (R-series), and establishment of the investigator as an expert consultant in the area of venous thromboembolic disease.